GPU Accelerated Real Time Rotation, Scale and Translation Invariant Image Registration Method
This paper presents highly optimized implementation of image registration method that is invariant to rotation scale and translation. Image registration method using FFT works with comparable accuracy as similar methods proposed in the literature, but practical applications seldom use this technique because of high computational requirement. However, this method is highly parallelizable and offloading it to the commodity graphics cards increases its performance drastically. We are proposing the parallel implementation of FFT based registration method on CUDA and OpenCL. Performance analysis of this implementation suggests that the parallel version can be used for real time image registration even for image size up to 2k x 2k. We have achieved significant speed up of up to 345x on NVIDIA GTX 580 using CUDA and up to 116x on AMD HD 6950 using OpenCL. Comparison of our implementation with other GPU based registration method reveals that our implementation performs better compared to other implementations.
KeywordsGPU Image Registration CUDA OpenCL Object Recognition
Unable to display preview. Download preview PDF.
- 1.Kuglin, C.D., Hines, D.C.: The Phase Correlation Image Alignment Method. In: Proceedings of the IEEE 1975 International Conference on Cybernetics and Society, New York, pp. 163–165 (1975)Google Scholar
- 2.DeCastro, E., Morandi, C.: Registration of Translated and Rotated Images Using Finite Fourier Transforms. IEEE Transactions on Pattern Analysis and Machine Intelligence, 700–703 (1995)Google Scholar
- 4.Sierra, I.: Geometric foundations for the uniqueness of the FFT- based Image Mosaicking, With The Application to Detecting Hidden Text in Web Images. M.S. Thesis, University of Texas at El Paso, El Paso, TX (131)Google Scholar
- 5.Rittavee, M.: Studies on log-polar Transformation For Image Registration and Improvements using Adaptive Sampling and Logarithmic Spiral. PhD. Thesis, Graduate School of Ohio State University, 5–40 (2009)Google Scholar
- 7.Ghafoor, A., Iqbal, R.N., Khan, S.: Robust Image Matching Algorithm. Video/Image Processing and Multimedia Communications, 155–160 (2003)Google Scholar
- 8.Lowe, D.G.: Distinctive Image Features from Scale-Invariant Keypoints. International Journal of Computer Vision, 91–110 (2004)Google Scholar
- 9.The CUDA C best practices guide, version 3.2. NVIDIA Corporation (August 2010), http://developer.download.nvidia.com/compute/cuda/32prod/toolkit/docs/CUDACBestPracticesGuide.pdf
- 10.Kirk, D.B., Hwu, W.W.: Programming massively parallel processors: A hands-on approach. Morgan Kaufmann, San Francisco (2010)Google Scholar
- 11.Martinez, M., Collet, A., Srinivasa, S.S.: MOPED: A scalable and low latency object recognition and pose estimation system. In: Robotics and Automation (ICRA), pp. 2043–2049 (2010)Google Scholar
- 12.Rublee, E., Rabaud, V., Konolige, K., Bradski, G.: ORB: An Efficient Alternative to SIFT or SURF. In: International Conference on Computer Vision (2011)Google Scholar
- 13.Chariot, A., Keriven, R.: GPU Boosted Online Image Matching. In: 19th International Conference on Pattern Recognition, pp. 1–4 (2008)Google Scholar